Field of the invention
[0001] The present invention refers to broadband wireless access networks, and more particularly
it concerns a connection-based scheduling method with differentiated service support
for hierarchical multi-hop relay networks. The invention can be used for instance
in networks based on IEEE standards 802.16x, which is one of the promising standards
where protocol elements are defined, worth of being considered when designing air
interfaces for new generation systems, i.e. beyond-3G (3rd Generation) and 4G (4th
Generation) systems.
[0002] In this respect, reference can be made to:
- IEEE 802.16-2004, IEEE Standard for Local and Metropolitan area networks - Part 16:
Air Interface for Fixed Wireless Access Systems, October 2004, and
- IEEE Std 802.16e-2005, Amendment to IEEE Standard for Local and Metropolitan Area
Networks - Part 16: Air Interface for Fixed Broadband Wireless Access Systems - Physical
and Medium Access Control Layers for Combined Fixed and Mobile Operation in Licensed
Bands, February 2006.
[0004] The document
ALAVI H S ET AL: "A Quality of Service Architecture for IEEE 802.16 Standards" COMMUNICATIONS,
2005 ASIA-PACIFIC CONFERENCE ON PERTH, WESTERN AUSTRALIA 03-05 OCT. 2005, PISCATAWAY,
NJ, USA, IEEE, 3 October 2005, pages 249-253, XP010860780 ISBN: 0-7803-9132-2, describes a architecture to support Quality of Services in IEEE 802.16 standards
as well as a design approach to implement such architecture.
Background of the Invention
[0005] The very high data rates envisioned for 4G wireless systems in reasonably large areas
do not appear to be feasible with the conventional cellular architecture due to two
basic reasons. First, the transmission rates envisioned for 4G systems are two orders
of magnitude higher than those of 3G systems, and it is well known that for a given
transmit power level, the symbol (and thus bit) energy decreases linearly with the
increasing transmission rate. Second, the spectrum that will be released for 4G systems
will almost certainly be located well above the 2 GHz band used by the 3G systems.
The radio propagation in these bands is significantly more vulnerable to non-line-of-sight
conditions, which is the typical mode of operation in today's urban cellular communications.
[0006] The brute-force solution to this problem is to significantly increase the density
of the base stations, resulting in considerably higher deployment costs, which would
only be feasible if the number of subscribers also increased at the same rate. This
seems unlikely to happen, the penetration of cellular phones and other mobile terminals
already being high in the developed countries. On the other hand, the same number
of subscribers will have a much higher demand in transmission rates. Since presumably
subscribers would not be willing to pay the same amount per data bit as for voice
bits, a drastic increase in the number of base stations does not seem therefore economically
justifiable.
[0007] However, fundamental enhancements are necessary for the very ambitious throughput
and coverage requirements of future systems. Towards this end, in addition to advanced
transmission techniques and co-located antenna technologies, some major modifications
in the wireless network architecture itself are required. The integration of multi-hop
relaying capability, by which an effective distribution and collection of signals
to and from the wireless users is entrusted not only to the base station but also
to other network elements (relays) is perhaps the most promising architectural upgrade
for extending the coverage of conventional (single-hop) wireless networks at reasonable
costs.
[0008] A multi-hop hierarchical relay network is a network where a base station is associated
with a plurality of Relay Nodes (RNs), arranged e.g. according to a logical tree structure,
and last-hop (or single-hop) connections are provided towards user terminals (UTs)
around each relay node. The multi-hop traffic is transmitted between the base station,
which is connected to a fixed backbone network, and the relay nodes that are strategically
placed. The last-hop traffic takes place between the relay node and a variable number
of user terminals.
[0009] The multi-hop technology allows enlarging the overall system coverage with low cost
infrastructures, since the relay nodes have a simpler structure and therefore are
cheaper than base stations. However, the task of ensuring Quality of Service (QoS)
requirements (throughput, delay, jitter, etc.) becomes more complex.
[0010] A resource request and allocation strategy at the Medium Access Control (MAC) level
keeping limited the end-to-end multi-hop delay has been proposed in our co-pending
European Patent Application No.
05425475.0, filed on 01.07.2005, entitled "Connection based scheduling method for hierarchical multi-hop wireless
networks extended to beyond 3G radio interface". That application represents the closest
prior art and claim 1 thereof recites (the parenthetical references to the Figures
are omitted):
"Method for controlling the access to a TDMA wireless channel from nodes deployed
as either a linear or tree topology network for multihop transmissions in uplink from
a requesting node to a centralized node and/or in downlink from the centralized node
towards an end node, including the steps of:
- issuing network topology information from the centralized to the other nodes;
- computing the amount of resources needed on each individual link between adjacent
nodes, by the transmitting node on that link;
releasing permissions, also called grants, for the use exclusive of TDMA channel for
a given time by the centralized node to each node along uplink and/or downlink multihop
path/s,
characterized in that said requesting node issues a cumulative request for the resources needed on each
link along the end-to-end path."
[0011] According to that strategy, the requests of resources for sending uplink flows from
relay nodes to the base station and/or downlink flows from base station to relay nodes
are computed by each requesting node for the end-to-end connection instead of being
computed only for the next link towards destination. This is just the meaning of "connection
based scheduling". This is made possible in networks with tree topology and centralized
scheduling where a request of resources is computed on individual links between two
adjacent nodes, and the network configuration is generally known to the requesting
nodes. In practice, each requesting node issues a cumulative request given by summing
up the same request for each link that separates the node from the base station (in
uplink) plus each link separating the base station from the destination node (in downlink).
The base station, in response to all cumulative requests, grants uplink and/or downlink
resources for each link. A grant is intended as an individual permission given to
the node for the use exclusive of the common resource (e.g. the TDMA radio channel)
for a fraction of time. The cumulative request/grant is made possible, e.g. in IEEE
802.16 networks, by the structure of the centralised control scheduling messages
[0012] This strategy, together with an order of transmission depending on the topology (in
uplink direction the node farthest form the base station transmits first and the node
closest to the base station transmits last, and in downlink direction transmission
occurs in the reverse order) guarantees that packets wait for being transmitted only
in the source relay nodes and not in the forwarding or transit relay nodes, and that
they are delivered to the destination within one frame once they are sent from the
source node. A further one-frame delay is to be considered in the average for the
last hop from/to the user terminal. Also, fairness in respect of the number of hops
and of the propagation direction (as shown by the delay curves reported in Figs. 16
and 17 of the application) is achieved.
[0013] However, this strategy does not take into account that a relay node generally handles
connections associated with services having different QoS requirements, such as, in
the simplest case, real time and non-real time services (e.g. to support both multimedia
and web browsing applications). A grant of resources determined on the basis of the
total traffic of the relay nodes can result ultimately in a risk of lack of resources
for real time traffic (or generally traffic with higher QoS requirements), especially
for nodes more distant from the base station: this results in turn in a degradation
of the QoS, especially in case traffic distribution among real time/non real time
services (or, generally, among different classes of service) at the different nodes
is non-uniform.
Summary of the Invention
[0014] Thus, it is an object of the present invention to provide a method of connection-based
scheduling that can result in a fair grant of resources among the different nodes
not only in terms of the overall traffic handled by a node but also in terms of the
different services to which the traffic refers.
[0015] According to the invention this object is achieved in that, in the presence of traffic
belonging to different classes of service, the cumulative request issued by a source
node for the resources needed on each link up to the destination node comprises the
concatenation of a plurality of individual cumulative requests each concerning traffic
belonging to a different class of service.
[0016] The method of the invention will also be referred to as "connection-based scheduling
with differentiated service support.
[0017] The grant of resources to a node is advantageously non-differentiated with respect
to the different classes of service, since the node knows the distribution of its
traffic demands and can suitably share the resources it has been granted.
[0018] Also the message relevant to the grant of resources reserved to said source node
could be concatenated at a transmitting node with the individual requests.
[0019] The invention starts from the following considerations. Considering by way of example
an IEEE 802.16 compliant network and assuming that each relay node is assigned a transmission
opportunity within each frame (and more particularly within the schedule control sub-frame)
for sending request and grant control messages (Mesh Centralized Scheduling - MSH-CSCH
- messages), 4 OFDM symbols are assigned to a relay node for the transmission of a
MSH-CSCH message. Therefore, assuming the most robust modulation and coding for the
control sub-frame, which corresponds to 24 bytes per OFDM symbol (QPSK 1/2), the available
resources are equal to 96 bytes. Now, it can be seen that, in the general case of
a tree network, the length of the MSH-CSCH message, in bytes, assuming no link updates
(see the IEEE Standard) is given by the following equation:
where NumRNs is the number of relay nodes within the multi-hop relay network and
is the number of relay nodes with a distance from the base station one hop higher
than the distance between the i-th relay node and the base station.
[0020] Relation (1) can be deduced from the MSH-CSCH message structure disclosed in the
standard (see Table 82). Reference can also be made to the paper "Performance Analysis
of IEEE 802.16a in Mesh Operation Mode", by S. Redana and M. Lott, Proceedings of
the 13th IST SUMMIT, Lyon, France, June 2004.
[0021] Taking into account that multi-hop networks will generally include a limited amount
of strategically located relay nodes, so that NumRNs is a rather small number, the
length of a conventional MSH-CSCH message is remarkably lower than the available resources.
The resources that are not exploited by the conventional MSH-CSCH message for that
relay node are thus exploited, according to the invention, for concatenating multiple
MSH-CSCH messages, in particular different request messages for resources to be allotted
to communications belonging to different classes of service.
Brief description of the drawings
[0022] Further objects, characteristics and advantages of the invention will become apparent
from the following description of preferred embodiments, given by way of non-limiting
example and illustrated in the accompanying drawings, in which:
- Fig. 1 is a schematic diagram of a multi-hop network;
- Fig. 2 is a diagram showing the request/grant profile with the connection-based algorithm;
- Fig. 3 is a graph of the mesh-frame structure according to standard IEEE 802.16;
- Fig. 4 is a graph showing the structure of the schedule control sub-frame according
to the present invention;
- Fig. 5 is a graph of the mean delay versus the total offered throughput in a multi-hop
network using the invention; and
- Fig. 6 is a graph of the mean delay versus the total offered throughput in a multi-hop
network not using the invention.
Description of the preferred embodiments
[0023] Referring to Fig. 1, there is schematically shown the architecture of a hierarchical
multi-hop relay network compliant with IEEE Standard 802.16. The network comprises
a plurality of relay nodes RN(x,y) that, from a logical point of view, are deployed
according to a tree topology, comprising x = 1...n branching levels and y = 1...N
i nodes per level. Each relay node has wireless access to the adjacent nodes and/or
to the base station, depending on its location along the branch. As indicated at the
bottom of the Figure, branching level x is x-hop spaced from the base station. A number
of user terminals UT are arranged around each relay node, as shown for relay node
RN(2,1), and have wireless access thereto. The whole of the relay nodes and the base
station forms a mesh structure supporting multi-hop links. A relay node and the user
terminals served by that node form a Point-to-Multipoint structure supporting single-hop
links. Multi-hop links are shown by solid lines and single hop links are shown by
dotted lines in the Figure.
[0024] We underline however that the tree configuration is a logical one, and that the physical
arrangement is strictly related with the characteristics of the area where the network
is deployed and, in general, will result in the provision of a limited number of strategically
located relay nodes.
[0025] The multiplexing of multi-hop and last-hop air interface can be performed according
to different concepts. The standard does not define the solution. The frequency domain
is a possible approach. The total frequency band is divided into two sub-bands: the
first subband is assigned for multi-hop and the second one for last-hop communication,
respectively. The Orthogonal Frequency Division Multiple Access (OFDMA) can be adopted
to split the available frequency band into two parts. Another approach is multiplexing
multi-hop and last-hop air interface in time domain. We consider a super frame as
combination of two frames: a frame assigned to multi-hop traffic and the other to
last-hop traffic.
[0026] For supporting the present invention, the network provides for a centralised, connection-based
scheduling. This means on the one hand that each node issues a cumulative request
for the end-to-end connection, containing resource requests not only for the next
hop but also for each link towards the destination, and, on the other hand, that the
base station gathers all cumulative requests and, in response, grants or allocates
uplink and/or downlink resources for each link involved in the connection.
[0027] Fig. 2 shows the profile of the requests/grants resulting from the application of
the above strategy to a path from BS to node RN(n,1). In the Figure, R
i,j and G
i,j with i, j =1...n, are the requests and the grants, respectively, on link j for the
connection (uplink and/or downlink) between BS and node RN(i,1). The Figure makes
it apparent that, due to the centralised management, the "population" of requests/grants
on each link increases as the distance from the base station decreases. It is to be
borne in mind that the Figure is only qualitative and that, even if the requests/grants
are shown by equally sized rectangles whatever the relay node and the link, this does
not imply any assumption on the amount of resources actually requested/granted.
[0028] Fig. 3 shows the frame organisation for the multi-hop operation mode in an IEEE 802.16
compliant network ("mesh operation mode", according to the standard), considering
for sake of simplicity and by way of example a branch of the tree structure with four
relay nodes denoted here as RN1, RN2, RN3, RN4.
[0029] A mesh frame consists of a control sub-frame and a data sub-frame, which are configured
by base station BS. According to the standard, two types of control sub-frames exist:
- Network Control sub-frame, used by BS to broadcast network information and by new
terminals that want to perform network entry.
- Schedule Control sub-frame, used by BS and the RNs to transmit requests and grants
for a new resource allocation within the data sub-frame. Only this second type of
control sub-frame is of interest for the invention and has been shown in the Figure.
[0030] A centralised scheduling is used and hence requests and grants are transmitted by
the base station and the relay nodes through MSH-CSCH messages. The Figure further
shows that each relay node RN1 - RN4 is assigned a transmission opportunity within
each frame for sending request and grant control messages. However, this is not mandatory.
The standard does not require that resources are assigned to each RN within one control
sub-frame: if a transmission opportunity for each relay node is not assumed within
one control sub-frame, the transmission order of request and grant messages does not
change but it is performed on multiple frames basis. However, since the grant message
does not reach each relay node in one frame, data collisions can occur due to non-updated
information on resource allocation in each relay node. Measures are to be studied
in order to avoid data collisions.
[0031] The data transmission occurs within the data sub-frame according to the last grant
message sent from the base station. Like in the co-pending European patent application,
the transmission order depends on the traffic direction (downlink/uplink) and is BS,
RN1, RN2, RN3 for downlink traffic and RN4, RN3, RN2, RN1 for uplink traffic. This
strategy guarantees that packets wait for being transmitted only in the source relay
node and not in the forwarding or transit relay nodes.
[0032] The invention aims at improving the connection-based scheduling so that the base
station, when granting resources to a relay node, can take into account that the traffic
handled by that node can relate to different classes of services with different QoS
requirements. To attain this goal , it is necessary that:
- each RN is able to classify and assign different classes of service, or equivalently
priorities, to different traffic flows (prioritisation);
- the BS is able to distinguish between service requests from each RN related to the
different classes of service and to establish a priority among the requests it receives
(differentiated service).
[0033] Packet classification is a standard function in the IEEE 802.16 compliant networks
considered in the described exemplary embodiment, and thus no problem exists in enabling
the relay nodes to perform it.
[0034] In order to allow a differentiated service support by the base station, according
to the invention more than one MSH-CSCH request messages are concatenated in the slot
of the scheduling control sub-frame assigned to a given relay node, each request concerning
a different class of service.
[0035] This is shown in Fig. 4 that refers, by way of non-limiting example only, to a coarse
classification of the traffic handled by a relay node RNi into "traffic with real-time
demands" (RT traffic) and "traffic without real-time demands" (non-real-time, nRT,
traffic). This can correspond for instance to distinguishing traffic relating to multimedia
and web browsing applications, respectively. In this example, when using the connection-based
scheduling with differentiated service support according to the invention, the slot
of the scheduling control sub-frame may contain the concatenation of an MSH-CSCH request
message requesting resources for the RT traffic and an MSH-CSCH request message requesting
resources for the nRT traffic. The requests for the different kinds of traffic (referred
to hereinafter as individual or partial requests) are still cumulative requests for
the resources needed on each link along the end-to-end path.
[0036] In case a node is to simultaneously forward both a resource request (in uplink) and
a grant (in downlink), also a MSH-CSCH grant message could be concatenated with the
individual request messages, since the transmission slot has generally sufficient
space to allow so.
[0037] Upon receiving the concatenated individual request messages, BS updates its perception
of resource needs for the links included in the message and, in the following frame,
computes and sends the grant message. Grants are calculated according to a BS-specific
policy that is not part of the present invention. For instance, considering the RT/nRT
example, BS could assign more grants on links where there are higher RT loads with
respect to other links with lower RT loads or could allot first resources to real-time
traffic and then allot the remaining resources, if any, to non-real traffic.
[0038] The actual grant on each link can be calculated for instance in the manner disclosed
in the above-mentioned patent application: the base station grants resources for each
link according to either a profile of grants equal to the profile of the requests
for that link if the whole amount of requested resources is below the maximum permissible
net throughput for the TDMA channel, or a profile lower than the profile of the requests
for that link, if the whole amount of requested resources is not below that maximum
permissible net throughput, wherein the lower profile is calculated through a normalisation
with respect to the ratio between said maximum permissible net throughput and the
whole amount of requested resources. Other strategies could result in BS favouring
only real time traffic, or favouring the farthest relay nodes, etc.
[0039] In any case a grant, as indicated in Fig. 4, is undifferentiated or aggregate, i.e.
it concerns the whole traffic handled by the node. Differentiation of the grants according
to the service classes is not necessary, since the relay node knows the composition
of its requests and can suitably allocate resources to the different service classes
without specific instructions from the base station.
[0040] Concatenation is made possible, as stated in the introduction of the specification,
because the planning strategies for multi-hop networks will lead to a limited number
of strategically located relay nodes, so that the length of a MSH-CSCH message is
much shorter than the length of the slot into which the message is to be inserted,
even if the most robust modulation is adopted for transmitting such message.
[0041] The only modification requested in the structure of the MSH-CSCH message defined
in the standard could be the addition of a "real time/non real time" flag to be set
to the proper value when the request/grant flag is 1 (request message). However, that
flag is not necessary if the order of concatenation of the requests for the different
classes of services is predefined.
[0042] Figs. 5 and 6 are graphs obtained by simulations carried out on the network with
four nodes considered in Fig. 3. The simulations have been performed considering the
IEEE 802.16 Point-to-MultiPoint air-interface specifications for the last hop from
a relay node to the user terminals. The graphs show the mean delay, expressed in number
of frames, vs. the total offered traffic (sum of the traffic to/from a node), assuming
a condition in which the aggregate traffic is almost the same at each node whereas
different nodes have different ratios of the real time to the non.real time traffic.
In particular, the load at RN1, RN2, RN3 is composed of 20% real-time traffic and
80% non real-time traffic and the load at RN4 is composed of 40% real-time traffic
and 60% non real-time traffic. Fig. 5 is obtained by using the invention and Fig.
6 without using the invention.
[0043] A comparison of the two graphs makes clearly apparent that, when BS is able to distinguish
between real-time and non real-time requests, it can grant a greater amount of radio
resources to RN4 than to the other relay nodes, to avoid that RN4 lacks resources
for the real time traffic. Thus a substantial fairness with respect to the number
of hops is obtained. On the contrary, if the differentiated service support is not
applied, all nodes would be allotted substantially the same resources and the resource
allocation would not be fair with respect to number of hops: actually, as shown, the
real time traffic aggregated at RN4 is penalised and reaches the saturation point
for lower values of offered traffic if compared with the other relay nodes.
[0044] Thus, the invention further enhances the improvements afforded by the connection-based
scheduling in terms of delay, obtaining fairness also among different kinds of connections.
Like the general principle of connection-based scheduling, the invention can be immediately
adopted in IEEE 802.16x networks with a very simple modification, or even no modification,
of the structure of the MSH-CSCH messages, as well as in beyond-3G systems with a
frame-based physical layer.
[0045] It is evident that the above description has been given by way of non-limiting example
and that changes and modifications are possible without departing from the scope of
the invention.
[0046] In particular, even if Fig. 4 shows the simple case of two types of traffic flows,
more than two request messages can be concatenated, if a finer classification is performed:
e.g. as many request messages as are the classes of service supported by the network
could be concatenated (for instance, the IEEE standard defines four classes of services).
In such case, the "real time/non real time" flag, if provided, will become a more
general "class of service" flag. Of course, the maximum number of messages that can
be concatenated depends on the number of relay nodes in the network, since the length
OH
MSH-CSCH of each message depends on such number, according to equation (1).
[0047] Moreover, even if the invention has been disclosed with particular reference to an
IEEE 802.16-compliant network, it can be adopted in any generic multi-hop wireless
network with the following features:
- tree topology with centralized scheduling;
- frame-based physical layer, wherein the MAC can align its scheduling intervals with
the underlying PHY framing.
1. A method of controlling the access to a common radio channel by nodes (BS, RN(1,1)...RN(n,Nn), RNi), in a wireless multihop communication network for multihop traffic from a
source node to a destination node wherein the network comprises a plurality of relay
nodes (RN(1,1)...RN(n,Nn), RNi) and a centralised node (BS) managing the access control, and wherein a relay
node (RN(1,1)...RN(n,Nn), RNi) having to access the channel is arranged to issue a cumulative request to
the centralised node summing up the request for the resources needed on each link
of a multi-hop path up to the destination node, characterised in that, in the presence of traffic belonging to different classes of service, said cumulative
request issued by said relay node (RN(1,1)...RN(n,Nn), RNi) comprises the concatenation of a plurality of individual cumulative requests
each concerning traffic belonging to a different class of service.
2. The method as claimed in claim 1, characterised in that said individual requests comprise at least an individual cumulative request for real
time traffic and an individual cumulative request for non-real time traffic.
3. The method as claimed in claim 1 or 2, characterised in that said cumulative requests comprise an individual cumulative request for each class
of service supported by the network.
4. The method as claimed in any of claims 1 to 4, characterised in that the centralised node (BS), in response to said plurality of concatenated requests
from a relay node (RN(1,1)...RN(n,Nn), RNi), is arranged to generate a single undifferentiated grant message for all classes
of service concerned by the individual requests.
5. The method as claimed in claim 4, characterised in that, at a relay node (RN(1,1)...RN(n,Nn), RNi) having to transmit a grant message besides a request message in downlink,
said grant message is concatenated with said plurality of individual requests.
6. The method as claimed in any of claims 1 to 5,
characterised in that said network is a network with the following features
- tree topology with centralized scheduling;
- frame-based physical layer, wherein the medium access control layer can align its
scheduling intervals with the framing of the underlying physical layer.
7. The method as claimed in claim 6, characterised in that said network is a network compliant with the Institute of Electrical and Electronics
Engineers, IEEE 802.16x set of standards.
8. The method as claimed in claim 7, characterised in that said individual requests are sent as Mesh Centralised Scheduling, MSH-CSCH, messages
including a supplementary flag for distinguishing the classes of service the individual
requests refer to.
9. The method as claimed in claim 7, characterised in that said individual cumulative requests are sent as Mesh Centralised Scheduling MSH-CSCH,
messages, where individual requests relevant to different classes of service are concatenated
in a predetermined and fixed order.
10. A relay node (RN(1,1)...RN(n,Nn), RNi), in a wireless multihop communication network for multihop traffic, having
to access a common radio channel, is arranged to issue a cumulative request to a centralised
node summing up the request for the resources needed on each link of a multi-hop path
up to a destination node, characterised in that, in the presence of traffic belonging to different classes of service, said cumulative
request issued by said relay node (RN(1,1)...RN(n,Nn), RNi) comprises the concatenation of a plurality of individual cumulative requests
each concerning traffic belonging to a different class of service.
1. Verfahren zum Steuern des Zugangs zu einem gemeinsamen Funkkanal durch Knoten (BS,
RN(1,1)...RN(n,Nn), RNi) in einem drahtlosen Multihop-Kommunikationsnetz für Multihop-Verkehr von einem
Quellenknoten zu einem Zielknoten, wobei das Netz mehrere Relayknoten (RN(1,1)...RN(n,Nn),RNi) und einen die Zugangssteuerung verwaltenden zentralisierten Knoten (BS) umfasst,
und wobei ein Relayknoten (RN(1,1)...RN(n,Nn), RNi), der auf den Kanal zugreifen muss, dafür ausgelegt ist, eine kumulative Anforderung
an den zentralisierten Knoten auszugeben, die die Anforderung der auf jeder Strecke
eines Multi-Hop-Pfads bis zu dem Zielknoten benötigten Betriebsmittel aufsummiert,
dadurch gekennzeichnet, dass bei Anwesenheit von zu verschiedenen Dienstklassen gehörendem Verkehr die durch den
Relayknoten (RN(1,1)...RN(n,Nn), RNi) ausgegebene kumulative Anforderung die Verkettung mehrerer individueller kumulativer
Anforderungen umfasst, die jeweils zu einer verschiedenen Dienstklasse gehörenden
Verkehr betreffen.
2. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass die individuellen Anforderungen mindestens eine individuelle kumulative Anforderung
für Echtzeitverkehr und eine individuelle kumulative Anforderung für Nicht-Echtzeitverkehr
umfassen.
3. Verfahren nach Anspruch 1 oder 2, dadurch gekennzeichnet, dass die kumulativen Anforderungen eine individuelle kumulative Anforderung für jede durch
das Netz unterstützte Dienstklasse umfassen.
4. Verfahren nach einem der Ansprüche 1 bis 4, dadurch gekennzeichnet, dass der zentralisierte Knoten (BS) dafür ausgelegt ist, als Reaktion auf die mehreren
verketteten Anforderungen von einem Relayknoten (RN(1,1)...RN(n,Nn), RNi), eine einzige undifferenzierte Gewährungsnachricht für alle Dienstklassen
zu erzeugen, die die individuellen Anforderungen betreffen.
5. Verfahren nach Anspruch 4, dadurch gekennzeichnet, dass in einem Relayknoten (RN(1,1)...RN(n,Nn), RNi), der neben einer Anforderungsnachricht in der Abwärtsstrecke eine Gewährungsnachricht
senden muss, die Gewährungsnachricht mit den mehreren individuellen Anforderungen
verkettet wird.
6. Verfahren nach einem der Ansprüche 1 bis 5,
dadurch gekennzeichnet, dass das Netz ein Netz mit den folgenden Merkmalen ist:
- Baumtopologie mit zentralisiertem Scheduling;
- Bitübertragungsschicht auf Frame-Basis, wobei die Mediumzugangskontrollschicht ihre
Scheduling-Intervalle mit dem Framing der zugrundeliegenden Bitübertragungsschicht
ausrichten kann.
7. Verfahren nach Anspruch 6, dadurch gekennzeichnet, dass das Netz ein Netz ist, das der Menge von Standards 802.16x des Institute of Electrical
and Electronics Engineers IEEE genügt.
8. Verfahren nach Anspruch 7, dadurch gekennzeichnet, dass die individuellen Anforderungen als Nachrichten des Mesh Centralised Scheduling MSH-CSCH
einschließlich eines Ergänzungs-Flag zur Unterscheidung der Dienstklassen, auf die
sich die individuellen Anforderungen beziehen, gesendet werden.
9. Verfahren nach Anspruch 7, dadurch gekennzeichnet, dass die individuellen kumulativen Anforderungen als Nachrichten des Mesh Centralised
Scheduling MSH-CSCH gesendet werden, wobei für verschiedene Dienstklassen relevante
individuelle Anforderungen in einer vorbestimmten und festen Reihenfolge verkettet
werden.
10. Relayknoten (RN(1,1)...RN(n,Nn), RNi) in einem drahtlosen Mulithop-Kommunikationsnetz für Multihop-Verkehr, der
auf einen gemeinsamen Funkkanal zugreifen muss, der dafür ausgelegt ist, eine kumulative
Anforderung an einen zentralisierten Knoten auszugeben, die die Anforderung der auf
jeder Strecke eines Multi-Hop-Pfads bis zu einem Zielknoten benötigten Betriebsmittel
aufsummiert, dadurch gekennzeichnet, dass bei Anwesenheit von Verkehr, der zu verschiedenen Dienstklassen gehört, die durch
den Relayknoten (RN(1,1)...RN(n,Nn), RNi) ausgegebene kumulative Anforderung die Verkettung mehrerer individueller kumulativer
Anforderungen umfasst, die jeweils zu einer verschiedenen Dienstklasse gehörenden
Verkehr betreffen.
1. Procédé pour commander l'accès de noeuds (BS, RN(1,1)...RN(n,Nn), RNi) à un canal radio commun dans un réseau de communication sans fil multisauts
pour un trafic multisauts d'un noeud source vers un noeud destinataire, dans lequel
le réseau comprend une pluralité de noeuds relais (RN(1,1)...RN(n,Nn), RNi) et un noeud centralisé (BS) gérant la commande d'accès et dans lequel un noeud
relais (RN(1,1)...RN(n,Nn), RNi) ayant à accéder au canal est adapté pour envoyer une demande cumulative au
noeud centralisé additionnant la demande en ressources nécessaires sur chaque liaison
d'un trajet multisauts jusqu'au noeud destinataire, caractérisé en ce que, en présence de trafic appartenant à différentes classes de services, ladite demande
cumulative envoyée par ledit noeud relais (RN(1,1)...RN(n,Nn), RNi) comprend la concaténation d'une pluralité de demandes cumulatives individuelles
concernant chacune le trafic appartenant à une classe de service différente.
2. Procédé selon la revendication 1, caractérisé en ce que lesdites demandes individuelles comprennent au moins une demande cumulative individuelle
pour le trafic en temps réel et une demande cumulative individuelle pour le trafic
non en temps réel.
3. Procédé selon la revendication 1 ou 2, caractérisé en ce que lesdites demandes cumulatives comprennent une demande cumulative individuelle pour
chaque classe de service supportée par le réseau.
4. Procédé selon l'une quelconque des revendications 1 à 4, caractérisé en ce que le noeud centralisé (BS), en réponse à ladite pluralité de demandes concaténées provenant
d'un noeud relais (RN(1,1)...RN(n,Nn), RNi), est adapté pour générer un unique message d'attribution indifférencié pour
toutes les classes de services concernées par les demandes individuelles.
5. Procédé selon la revendication 4, caractérisé en ce que, au niveau d'un noeud relais (RN(1,1)...RN(n,Nn), RNi) ayant à transmettre un message d'attribution en plus d'un message de demande
dans le sens descendant, ledit message d'attribution est concaténé avec ladite pluralité
de demandes individuelles.
6. Procédé selon l'une quelconque des revendications 1 à 5,
caractérisé en ce que ledit réseau est un réseau possédant les caractéristiques suivantes
- topologie en arbre à programmation centralisée ;
- couche physique basée sur la trame, la couche de commande d'accès au support pouvant
aligner ses intervalles de programmation avec le tramage de la couche physique sous-jacente.
7. Procédé selon la revendication 6, caractérisé en ce que ledit réseau est un réseau conforme à l'ensemble de normes IEEE 802.16x de l'Institut
des ingénieurs du domaine électrique et électronique (IEEE).
8. Procédé selon la revendication 7, caractérisé en ce que lesdites demandes individuelles sont envoyées en tant que messages de programmation
centralisée par maillage, MSH-CSCH, incluant un drapeau supplémentaire pour distinguer
les classes de services auxquelles se réfèrent les demandes individuelles.
9. Procédé selon la revendication 7, caractérisé en ce que lesdites demandes cumulatives individuelles sont envoyées en tant que messages de
programmation centralisée par maillage, MSH-CSCH, les demandes individuelles relatives
à différents classes de services étant concaténées dans un ordre fixe prédéterminé.
10. Un noeud relais (RN(1,1)...RN(n,Nn), RNi) dans un réseau de communication sans fil multisauts pour un trafic multisauts
ayant à accéder à un canal radio commun est adapté pour envoyer une demande cumulative
à un noeud centralisé additionnant la demande en ressources nécessaires sur chaque
liaison d'un trajet multisauts jusqu'à un noeud destinataire, caractérisé en ce que, en présence de trafic appartenant à différentes classes de services, ladite demande
cumulative envoyée par ledit noeud relais (RN(1,1)...RN(n,Nn), RNi) comprend la concaténation d'une pluralité de demandes cumulatives individuelles
concernant chacune le trafic appartenant à une classe de service différente.